Streaming ML Insights
The conversation highlights the critical distinction between batch and streaming machine learning, emphasizing the importance of recency in model evaluation. Shreya argues that blindly applying traditional train-validation-test splits can lead to outdated models that fail to reflect current data trends. The need for continuous retraining and thoughtful validation practices is underscored, as today’s predictions hold more weight than those from the past.In this clip
From this podcast
The Gradient
Shreya Shankar: Machine Learning in the Real World
Related Questions
What is the main topic of the clip Real-time vs. Batch from the episode Analyzing the Google Paper on Continuous Delivery in ML // Part 4 // MLOps Coffee Sessions #17?
What metrics are important in evaluating artificial intelligence in the context of the episode Analyzing the Google Paper on Continuous Delivery in ML // Part 4 // MLOps Coffee Sessions #17 and the clip Model Validation Challenges?
What metrics are important in evaluating artificial intelligence in the context of the episode "Analyzing the Google Paper on Continuous Delivery in ML // Part 4 // MLOps Coffee Sessions #17" and the clip "Model Validation Challenges"?